Exome sequencing and targeted gene panels: a simulated comparison of diagnostic yield using data from 158 patients with rare diseases

Abstract Next-generation sequencing (NGS) has altered clinical genetic testing by widening the access to molecular diagnosis of genetically determined rare diseases. However, physicians may face difficulties selecting the best diagnostic approach. Our goal is to estimate the rate of possible molecular diagnoses missed by different targeted gene panels using data from a cohort of patients with rare genetic diseases diagnosed with exome sequencing (ES). For this purpose, we simulated a comparison between different targeted gene panels and ES: the list of genes harboring clinically relevant variants from 158 patients was used to estimate the theoretical rate of diagnoses missed by NGS panels from 53 different NGS panels from eight different laboratories. Panels presented a mean rate of missed diagnoses of 64% (range 14%-100%) compared to ES, representing an average predicted sensitivity of 36%. Metabolic abnormalities represented the group with highest mean of missed diagnoses (86%), while seizure represented the group with lowest mean (46%). Focused gene panels are restricted in covering select sets of genes implicated in specific diseases and they may miss molecular diagnoses of rare diseases compared to ES. However, their role in genetic diagnosis remains important especially for well-known genetic diseases with established genetic locus heterogeneity.

摘要 下一代测序（Next-generation Sequencing, NGS）通过拓展遗传性罕见病分子诊断的可及性，重塑了临床基因检测的整体格局。然而临床医师常面临最优诊断方案的选择困境。本研究旨在依托经外显子组测序（Exome Sequencing, ES）确诊的罕见遗传病患者队列数据，评估不同靶向基因检测面板（targeted gene panels）可能漏诊的分子诊断比例。为此，我们模拟了不同靶向基因检测面板与外显子组测序的对比分析：纳入158例患者携带临床意义变异的基因列表，以此估算8家实验室提供的53种不同NGS面板的理论漏诊率。相较于外显子组测序，各类检测面板的平均漏诊率为64%（区间14%~100%），对应平均预测灵敏度为36%。亚组分析显示，以代谢异常为表型的患者亚组平均漏诊率最高（86%），而以癫痫发作为表型的亚组平均漏诊率最低（46%）。靶向聚焦的基因检测面板仅覆盖特定疾病相关的有限基因集，相较于外显子组测序更易漏诊罕见病的分子诊断。但这类检测面板在基因诊断中仍具有重要价值，尤其适用于已明确遗传位点异质性的常见遗传病。